Graphite1.0-4B / run_summary.json
Starred
Training in progress, step 250
f2661f1 verified
raw
history blame
556 Bytes
{
"model_id": "Qwen/Qwen3.5-4B-Base",
"train_seed": 3407,
"packing": false,
"eval_packing": false,
"completion_only_loss": false,
"assistant_only_loss": false,
"max_seq_length": 2048,
"learning_rate": 0.0001,
"lr_scheduler_type": "cosine",
"smoke_test": false,
"smoke_test_max_steps": 30,
"hub_push_enabled": true,
"hub_model_id": "Starred09/obsidian-critic-qwen35-4b-base-lora",
"hub_strategy": "checkpoint",
"resume_from_checkpoint": false,
"train_examples_after_filter": 36081,
"validation_examples_after_filter": 363
}